Revue europenne de psychologie applique 61 (2011) 111

Original article

Music listening and emotional well-being in adolescence: A person- and variable-oriented study
coute de la musique et bien-tre motionnel ladolescence : une tude centre sur les personnes et sur les variables
D. Miranda , P. Gaudreau
School of Psychology, University of Ottawa, 200, Lees avenue, Ottawa, Ontario, K1N 6N5, Canada

a r t i c l e

i n f o

a b s t r a c t
The principal aim of this study was to determine if different proles (types) of emotional reactions following music listening (happiness and sadness) characterized different levels of emotional well-being (i.e., positive and negative affects) in adolescence. The secondary aim was to examine relationships between social congruence in music tastes with friends or parents (i.e., sharing similar music tastes and having fewer conicts about music) and emotional well-being in adolescence. This studys sample was composed of 316 adolescents (M = 15.32 and S.D. = 0.90 years of age; 172 girls and 144 boys). Cluster analysis identied three proles: (1) emotionally-negative listeners (medium happiness and higher sadness); (2) emotionally-limited listeners (lower happiness and lower sadness); (3) emotionally-positive listeners (higher happiness and lower sadness). Results indicated that emotionally-negative listeners had less emotional well-being, that emotionally-positive listeners had more emotional well-being, and that social congruence in music tastes with both friends and parents were related to more emotional well-being. 2010 Elsevier Masson SAS. All rights reserved.

Article history: Received 23 March 2010 Received in revised form 23 October 2010 Accepted 25 October 2010 Keywords: Music Adolescence Emotional well-being Social congruence Cluster analysis

r s u m
Mots cls : Musique Adolescence Bien-tre motionnel Congruence sociale Analyse par regroupements

Lobjectif primaire de cette tude tait de dterminer si diffrents prols (types) de ractions motionnelles suite lcoute de la musique (bonheur et tristesse) caractrisent diffrents niveaux de bien-tre motionnel (i.e., affects positifs et ngatifs) ladolescence. Lobjectif secondaire tait dexaminer les relations entre la congruence sociale des gots musicaux avec les amis ou les parents (i.e., partager des gots musicaux similaires et avoir moins de conits propos de la musique) et le bien-tre motionnel ladolescence. Lchantillon de cette tude tait compos de 316 adolescents (M = 15.32 et ET = 0,90 ans ; 172 lles et 144 garc ons). Une analyse par regroupements (cluster analysis) a identi trois prols : (1) auditeurs motionnellement ngatifs (bonheur moyen et tristesse leve) ; (2) auditeurs motionnellement limits (bonheur faible et tristesse faible) ; (3) auditeurs motionnellement positifs (bonheur lev et tristesse faible). Les rsultats ont indiqu que les auditeurs motionnellement ngatifs avaient moins de bien-tre motionnel, que les auditeurs motionnellement positifs avaient plus de bien-tre motionnel et que la congruence sociale des gots musicaux avec les amis et les parents tait associe plus de bien-tre motionnel. 2010 Elsevier Masson SAS. Tous droits rservs.

1. Introduction As a scientic endeavour, studying the interface between developmental and music psychology is an integrative effort to understand how musically-related behaviours, emotions, and cognitions evolve across the life course and how they interact positively and/or negatively with psychosocial adaptation. The developmental study of music is thus an interdisciplinary research

Corresponding author. Co-corresponding author. E-mail addresses: dave.miranda@uottawa.ca (D. Miranda), pgaudrea@uottawa.ca (P. Gaudreau). 1162-9088/$ see front matter 2010 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.erap.2010.10.002

D. Miranda, P. Gaudreau / Revue europenne de psychologie applique 61 (2011) 111

theme that cuts across several research elds, such as psychology, education, and cognitive neuroscience (e.g., Hargreaves, 1986; McPherson, 2006; Peretz and Zatorre, 2003). In regard to the developmental implications of music listening, adolescence has been thought of being particularly meaningful, notably because motivation for music listening can reach its climax during the second decade of life (Larson, 1995; North et al., 2000; Zillmann and Gan, 1997). Improving our knowledge about the developmental implications of music listening in adolescence provides novel opportunities to better understand distinctive psychosocial needs and realities of adolescents. For instance, it has provided an original outlook to further understand developmental constructs established by mainstream psychology, such as personality, coping, or friendship (e.g., Delsing et al., 2008; Miranda and Claes, 2008, 2009; Rentfrow and Gosling, 2003; Selfhout et al., 2009). Ultimately, knowledge about the developmental roles of music can be used to optimize evidence-based prevention and intervention programs insofar as making them more developmentally attuned to the lifestyle of contemporary adolescents (see Lemieux et al., 2008). Moreover, developmentally attuned music therapies for adolescents can intervene on and inform about key aspects of individual and social youth development, such as identity formation, resilience, connectedness, and competence (McFerran, 2010). Music listening is usually a pleasurable, gratifying, and valuable daily experience throughout the life course. Nonetheless, in terms of psychosocial adaptation, music listening in adolescence can also relate (either as a protective or risk factor) to various indicators of psychopathology. Therefore, it is also possible to talk about a developmental psychopathology of music listening as a research topic implicitly shared by several studies in adolescence. For instance, recent studies have shown complex longitudinal links between different music listening behaviours and externalizing problem behaviours in adolescence, such as aggression (Selfhout et al., 2008). Recent studies have also identied complex longitudinal relationships between different music listening behaviours and internalizing symptoms in adolescence, such as depression (Miranda and Claes, 2008). Recent literature reviews have also scrutinized developmental issues surrounding music in adolescence (see North and Hargreaves, 2008). Overall, such developmental studies and reviews (whether on externalizing or internalizing symptoms) deepen our understanding of adaptive (or maladaptive) music listening as it relates to the negative end of the continuum of mental health: psychopathology. Conversely, the present study strived to improve our knowledge about the relationships between music listening in adolescence and the positive end of the continuum of mental health: subjective well-being. Subjective well-being is colloquially referred to as happiness, and it can be scientically dened as the optimal presence of higher levels of positive affect (pleasant emotions), lower levels of negative affect (unpleasant emotions), and higher levels of life satisfaction (see Diener, 2000). As such, this study was inspired by the gradual shift of psychology from its traditional emphasis on psychopathology to placing a stronger consideration on the study of positive human development. The recent movements of positive psychology and of psychology of positive youth development are representative examples of this scientic trend (Larson, 2000; Seligman and Csikszentmihalyi, 2000). Therefore, it is within a positively-oriented developmental research that the present study undertook to investigate links between music listening and adolescents emotional well-being, a construct consisting of higher positive affect and lower negative affect. In fact, it seemed especially pertinent to focus on the emotional component of subjective well-being (i.e., positive and negative affects) as music listening is profoundly tied to both positive and negative emotional experiences (Juslin and Sloboda, 2010).

Lastly, the developmental roles of music listening in adolescence can also be organized by two overreaching motivation systems. The rst motivation system is responsible for the satisfaction of individual needs (e.g., emotion regulation, aesthetic appreciation, coping) whereas the second is responsible for the satisfaction of social needs (e.g., social identity, relationships with peers, membership to musical subcultures) (Bakagiannis and Tarrant, 2006; Miranda and Claes, 2009). In this study, this framework was used to account for individual experiences (emotional reactions from music listening) and social experiences (social congruence in music tastes with friends or parents) involved in the music listening of adolescents. 1.1. Individual experiences: emotions from music listening and emotional well-being Research suggests that positive and negative affects are two fundamental and distinct dimensions structuring emotions (Watson and Tellegen, 1985). For instance, it is possible to have simultaneously negative emotions and positive emotions towards (or as a result) of any given experience what is colloquially referred to as having mixed feelings. In other words, there is mounting empirical evidence that different levels of negative emotions and positive emotions can be experienced simultaneously in each and every person (Watson and Clark, 1997). In parallel, two of the basic emotions that are activated by music listening are happiness (positive) and sadness (negative) (Kallinen, 2005). In this regard, the eld of cognitive neuroscience provides compelling biological evidence for the direct inuence of music on human emotions (see Juslin and Sloboda, 2010; Peretz and Zatorre, 2003). Additionally, recent experimental studies indicated that both positive and negative emotional responses can coexist within the same music listening experience (Hunter et al., 2008). Lastly, follow-up studies in natural settings using an experience sampling method that provided a random sample of everyday emotional states corroborated that, on average and to various degrees, both happiness and sadness are experienced from everyday life music listening (Juslin et al., 2008). In other words, different levels of negative emotions and positive emotions can be experienced simultaneously when one person listens to music. Can different patterns of this simultaneous experience of happiness and sadness from music listening be associated with adolescents emotional well-being? Firstly, this seems likely if one considers the importance that many adolescents give to music and the great amount of music listening they can experience on a daily basis (Roberts et al., 2004). Secondly, one should also add to this consideration that daily music listening conveys mild and cumulative emotional experiences that can become signicant over time (Sloboda, 2010). However, to the best of our knowledge, research has rarely studied relationships between basic emotional reactions from music listening and emotional well-being in adolescence, as the most related research on music in adolescence has prioritized the study of health risks, social issues, and psychopathology. For instance, in terms of externalizing problem behaviours, Roberts et al. (1998b) found that negative emotions experienced from music listening was a robust predictor of health risk behaviours (e.g., drug use, violence) in a sample of adolescents. In terms of internalizing problem behaviours, Martin et al. (1993) have reported that experiencing more sadness as a result of music listening can also be associated with higher levels of depression. In sum, a generally positive emotional experience from music listening could be part of, or play a role in, the normative psychosocial development of adolescents. The current literature provides some converging evidences supporting this assumption. Firstly, among main reasons, adolescents report listening to music because it provides positive and enjoyable experiences (North et al., 2000;

D. Miranda, P. Gaudreau / Revue europenne de psychologie applique 61 (2011) 111

Tarrant et al., 2000). Secondly, happiness usually occurs more often than sadness during music listening (Juslin et al., 2008). Thirdly, adolescents who experience more negative emotions from music listening have poorer mental health (Martin et al., 1993; Roberts et al., 1998b). Hence, it is conceivable that experiencing more happiness (and less sadness) from music listening could be linked to more emotional well-being in youth. 1.2. Social experiences: music listening and social congruence In adolescence, social identity and social relationships undergo tremendous changes as adolescents gradually shift their relationships from parents to peers (Brown and Klute, 2003; Brown and Larson, 2009). As such, music listening behaviours seem to follow (or facilitate) this developmental trend towards an increase in peer-related relationships (Zillmann and Gan, 1997). Understandably, research on music listening in adolescence has focused on its contribution to subcultures, peer afliation, and friendship formation (e.g., Miranda and Claes, 2009; Selfhout et al., 2009). Therefore, it seems understandable that research on the developmental role of music listening in adolescence has not considered parents as much as peer relations. Still, from a developmental perspective, the lack of studies on parents and music listening in adolescence represents a research caveat given that relationships with parents remain extremely important throughout adolescence (Kerr et al., 2003). In general, the role of parents in musical youth development has been investigated by only a limited number of studies (McPherson, 2009). Hence, in order to help ll this research gap, this study attempted to consider social congruence in music tastes that are harmoniously shared not only with friends, but also with parents. Congruence, as applied to interpersonal relationships, can refer to the tting of experiences between the self and social experiences, for instance in terms of satisfactory reciprocal communications between individuals, a phenomenon that can promote actualization and psychological adjustment (Rogers, 1961). In this study, social congruence in music tastes, with friends or parents, is broadly dened as sharing more similar music tastes and experiencing fewer interpersonal conicts regarding music tastes with either social milieu. Experiencing social congruence from sharing more similar music tastes with friends or parents could then be a contributory factor to development, given that maintaining good relationships with peers, friends, and parents is crucial in adolescent development (Hartup and Stevens, 1997; Kerr et al., 2003; Youniss and Smollar, 1985). Thus, social congruence in music tastes with friends or parents could be related to more emotional wellbeing in adolescence. 1.3. Gender differences in music listening: emotional and social behaviours Gender differences are present in the emotional and social aspects of music listening during adolescence, such as girls experiencing more emotional regulation from music listening (Miranda and Claes, 2007, 2008, 2009; North et al., 2000). For instance, signicant gender differences were observed by Miranda and Claes (2009) using self-report measures of music listening behaviours within a correlational research design among 418 adolescents. They have found that adolescent girls were using emotional coping by music listening signicantly more often than adolescent boys. North et al. (2000) have also observed signicant gender differences using self-report measures of music listening behaviours within a correlational research design among 2465 adolescents. At the emotional level, their results have indicated that adolescents reported using music listening behaviours as means to full emotional needs. However, girls gave signicantly more importance to

this emotional function of music compared to boys (North et al., 2000). Moreover, at the social level, their results have also indicated that adolescents reported using music listening behaviours as ways to manage their social identity in the presence of other people. Boys, however, attached signicantly more importance to this social purpose of music compared to girls (North et al., 2000). 1.4. Objectives and hypotheses 1.4.1. Main objective The principal aim of this study was to determine if different proles (types) of emotional reactions following music listening (happiness and sadness) could characterize different levels of emotional well-being (positive and negative affects). The literature has enabled the proposal of two hypotheses, which take into account that happiness and sadness can coexist simultaneously in music listening experiences (Hunter et al., 2008; Juslin et al., 2008). Hypothesis 1. The rst hypothesis suggested that most adolescents would be characterized by experiencing more happiness (over sadness) from music listening because such pattern could be part of their normative development. This was suggested because music listening usually provides positive emotional experiences (Juslin et al., 2008; North et al., 2000; Tarrant et al., 2000), whereas those adolescents who experience more negative emotions from music listening may have poorer mental health (Martin et al., 1993; Roberts et al., 1998b). Hypothesis 2. The second hypothesis proposed that adolescents who experience more happiness (over sadness) from music listening would report a better emotional well-being than those experiencing more sadness (over happiness) from music listening. This proposition was supported by three basic features of music listening (Juslin et al., 2008; Larson, 1995; North et al., 2000; Roberts et al., 2004; Sloboda, 2010; Zillmann and Gan, 1997): it habitually provides positive emotions; it is very important for many adolescents; it can have cumulative emotional inuences. The rst argument outlines the main valence that the inuence of positive emotional trade-offs from music listening may have on emotional well-being. Positive emotions are known to be intrinsic to emotional well-being. Therefore, if music primarily creates positive emotional trade-offs, in turn, music may contribute to better emotional well-being. The second argument underscores the accessibility of such positive emotional trade-offs from music listening. Many adolescents listen to a lot of their favourite music on a daily basis. Thus, this listening tendency may multiply the daily occasions for music to have emotional impacts. The third argument underlines the strength that the inuence of positive emotional trade-offs from music listening may have on emotional well-being. Music may only gradually inuence emotional well-being, but such accumulation may still be partially and signicantly captured by our aggregated self-report measures of emotional states. Moreover, it was considered that possible associations between musical emotional trade-offs and emotional well-being could be moderated by gender, given that gender differences are present in music listening during adolescence, such as girls experiencing more emotional regulation from music listening (Miranda and Claes, 2007, 2008, 2009; North et al., 2000). 1.4.2. Secondary objective The secondary aim of this study was to verify the presence of signicant relationships between social congruence in music tastes with friends or parents (i.e., sharing similar music tastes and having fewer conicts about music) and emotional well-being.

D. Miranda, P. Gaudreau / Revue europenne de psychologie applique 61 (2011) 111

Hypothesis 3. The third hypothesis predicted that social congruence in music tastes with friends and parents would be related to more emotional well-being. This was grounded in mainstream research, which indicates that the development of good relationships with both peers and parents is crucial for adolescent development (Hartup and Stevens, 1997; Kerr et al., 2003; Youniss and Smollar, 1985). Once more, it was veried if these relationships applied for both girls and boys because of the gender differences in music listening during adolescence, such as boys attaching more importance to managing their social identity through musical behaviours in the presence of other people (Miranda and Claes, 2007, 2008; North et al., 2000). 2. Method 2.1. Sample and procedure This study is part of a broader research program that investigates the developmental roles of music listening in adolescence. The sample available for this study was composed of 329 adolescents (M = 15.34 and S.D. = 0.90 years of age; 179 girls and 150 boys). Participants were French-speaking students who attended grades 9 to 11 of a large public French high school in Montreal (Canada). After being informed of the research, participants voluntarily consented to lling out a self-report questionnaire (written in French) for 30 minutes of their regular classes. This sample was also utilized in a previous study (Miranda and Claes, 2007). Box plots were used to identify all possible outliers with extreme scores on each of the six variables in this study. In a box plot, the score of an outlier falls at an extreme distance (more than 1.5 times the interquartile range) from either below the rst quartile or above the third quartile. Accordingly, using box plots, four multivariate outliers were identied as they had extreme scores on two or more variables. They were thus discarded from the main analyses. Additionally, again using box plots, nine univariate outliers on happiness and sadness from music listening were also discarded from the main analyses, as these two variables were later used for cluster analysis a statistical procedure that can be particularly sensitive to outliers. Consequently, the nal sample of this study was composed of 316 adolescents (M = 15.32 and S.D. = 0.90 years of age; 172 girls and 144 boys). This nal sample had an equivalent proportion of girls and boys ( 2 = 2.48, P > 0.05). Many of these adolescents had few siblings, for instance, 44.3% had no brothers and 38.3% had no sisters. Adolescents parents were either married (77.8%), separated/divorced (16.1%), or other (6.1%). Most adolescents parents were working in full-time jobs (fathers = 87.7%; mothers = 58.5%) and had completed at least their high school education (fathers = 89.9%; mothers = 92.7%). Adolescents had diversied cultural backgrounds as many of their parents were immigrants (fathers = 64.3%; mothers = 57.9%). 2.2. Measures 2.2.1. Emotional well-being A French version (Gaudreau et al., 2006) of the Positive Affect Negative Affect Schedule (PANAS) (Watson et al., 1988) was utilized to assess adolescents current emotional well-being. Twenty adjectives were rated on a ve-point scale (not at all or very slightly to extremely). Ten adjectives pertained to positive affect (e.g., enthusiastic, inspired) and ten adjectives to negative affect (e.g., upset, nervous). The PANAS has shown good internal consistency, test-retest reliability, factorial validity, and external validity (Watson et al., 1988). It is not a diagnostic tool of mood disorders, but rather a prominent self-report measure of basic moods.

As such, however, experiencing low positive affect and high negative affect can be associated with having more symptoms of mood disorders (e.g., distress, depression, anxiety; Watson et al., 1988). Nowadays, the PANAS is one of the most frequently used psychometric measure of emotional states/traits, emotional adjustment, and emotional well-being in major elds of psychological research, including but not restricted to: developmental psychology in adolescence (Gaudreau et al., 2009), personality psychology (Wilson and Gullone, 1999), social psychology (Koestner et al., 2002), sport psychology (Gaudreau and Blondin, 2004), psychology of emotion (Harmon-Jones et al., 2009), and positive psychology of music (Laukka, 2007). The French version of the PANAS had previously undergone a thorough double back-translation, it had shown adequate internal consistency (Cronbachs alphas), and factorial validity (conrmatory factor analysis) across different samples (Gaudreau et al., 2006). Recent studies on the PANAS have further suggested that a three-factor model (e.g., positive affect, afraid, and upset) may provide a better t to the data than the usual two-factor model (positive and negative affects; Gaudreau et al., 2006). Nevertheless, for this exploratory study, we favoured the usual two-factor model (positive and negative affects), given that it is still the classical model both in psychology (Watson and Clark, 1997) and previous research on music and emotions in adolescence (Roberts et al., 1998b). In this study, in terms of Cronbachs alpha coefcients, both positive ( = 0.83) and negative affect ( = 0.82) subscales presented good internal consistency. 2.2.2. Emotional state following music listening Happiness resulting from music listening was measured through an exploratory single-item measure: I feel happy after listening to my favourite music. . .. Sadness resulting from music listening was measured by another exploratory single-item measure: I feel unhappy after listening to my favourite music. . .. These items originated from the psychometric work of Lacourse (1999) on music in adolescence. Both items were rated on a fourpoint scale (1 = not at all; 4 = a lot). 2.2.3. Social congruence in music tastes (friends or parents) Social congruence in music tastes with friends was measured with two exploratory items: Does your friends listen to the same music as you?; Does it happen that you quarrel with your friends over the music you prefer?. Social congruence in music tastes with parents was measured by two analogous exploratory items: Does your parents listen to the same music as you?; Does it happen that you quarrel with your parents over the music you prefer?. All four items were rated on a four-point scale (1 = never; 4 = very often). 3. Results 3.1. Correlations and descriptive statistics Table 1 presents the correlations (Spearmans rho) and descriptive statistics for the variables. Essentially, these results were useful in verifying the absence of signicant correlations between positive affect and negative affect, but also the absence of signicant correlations between happiness and sadness from music listening. Hence, these results have indicated that positive affect and negative affect were orthogonal dimensions, which is in accordance with the liter-

D. Miranda, P. Gaudreau / Revue europenne de psychologie applique 61 (2011) 111 Table 1 Correlations (rs ) and descriptive statistics (n = 316). Variables 1. Positive affect 2. Negative affect 3. Happiness from music listening 4. Sadness from music listening 5. Social congruence music taste/friends 6. Social congruence music taste/parents 1 2 0.03 3 0.11 0.02 4 0.12* 0.27*** 0.01 5 0.02 0.07 0.11 0.13* 6 0.13* 0.20*** 0.07 0.09 0.05 M (S.D.)

M (S.D.) Total (n = 316) Girls (n = 172) Boys (n = 144) t-tests (girls-boys) * P < 0.05, ** P < 0.01, *** P < 0.001. Table 2 Descriptive statistics of this studys three music listeners proles (n = 316). 31.60 (6.66) 31.37 (6.50) 31.87 (6.87) 0.66

M (S.D.) 21.83 (6.33) 21.95 (6.11) 21.68 (6.59) 0.37

M (S.D.) 3.16 (0.81) 3.27 (0.71) 3.03 (0.91) 2.70**

M (S.D.) 1.21 (0.44) 1.22 (0.44) 1.19 (0.45) 0.53

M (S.D.) 6.70 (1.12) 6.71 (1.17) 6.68 (1.06) 0.28

5.13 (1.21) 5.15 (1.23) 5.10 (1.20) 0.30

Emotionally-negative listeners (ENL)

Emotionally-limited listeners (ELL)

Emotionally-positive listeners (EPL) M (S.D.) 4.00 (0.00) 1.00 (0.00) 1.03 (0.00) 0.47 (0.00) n (%) 100 (31.6) 57 (33.1) 84 (58.3) 1.96

M (S.D.) Happiness from music listening (centroids) Sadness from music listening (centroids) Happiness from music listening (z-scores) Sadness from music listening (z-scores) 3.20 (0.79) 2.08 (0.28) 0.04 (0.97) 1.96 (0.62) n (%) Total sample Girls subsample Boys subsample 2 (girls-boys) 61 (19.3) 36 (20.9) 25 (17.4) 1.98
2

M (S.D.) 2.61 (0.59) 1.00 (0.00) 0.68 (0.72) 0.47 (0.00) n (%) 155 (49.1) 79 (45.9) 35 (24.3) 0.06 = 0.72, P > 0.05), but not in boys (
2

Proportions of the three proles in the total sample is equivalent in girls (

= 50.88, P < 0.001). * P < 0.05, ** P < 0.01, *** P < 0.001.

ature on emotions (Watson and Tellegen, 1985). These results have also shown that happiness and sadness from music listening were also orthogonal dimensions, which is in conformity with the literature on music and emotions (Hunter et al., 2008; Juslin et al., 2008). As for gender differences, t-tests indicated that girls had higher levels of happiness from music listening than boys (Table 1). This result suggests that these adolescent girls and boys may have differed in terms of the intensity of their emotional reaction to music. This would be compatible with the variable-oriented literature that showed gender differences in terms of emotional regulation involving music listening in youth (Miranda and Claes, 2007, 2008, 2009; North et al., 2000). 3.2. Emotional proles of music listeners The principal aim of this study required to account for the fact that happiness and sadness can coexist simultaneously in the experience of music listening (Hunter et al., 2008; Juslin et al., 2008). Thus, a person-oriented approach was adopted in order to consider the typical simultaneous organization of these two variables within each individual. In this study, a cluster analysis enabled to identify a number of homogeneous and mutually exclusive proles (types or subgroups) of individuals from the combination of their simultaneous scores of happiness and sadness resulting from music listening. Such person-oriented approaches are growing in psychology and have already been fruitfully used in psychology of music (e.g., Mulder et al., 2007), in developmental and personality psychology (e.g., Morizot and Le Blanc, 2005), and in research on subjective well-being (e.g., Busseri et al., 2009). Hence, a hierarchical cluster analysis was conducted (using Wards method and squared Euclidean distance) in order to categorize participants into a denite number of homogeneous

and mutually exclusive subgroups, also labeled types or proles in the cluster analytical parlance (Aldenderfer and Blasheld, 1984). Results indicated the possibility of clustering participants in either three or four subgroups, both solutions accounting for the simultaneous dual pattern of scores of happiness and sadness from music listening within each participant. The threecluster solution explained 86% of variance, which represented an increment of 11% of explained variance over a two-cluster solution. The four-cluster solution explained 92% of variance, which represented an increment of only 6% of explained variance over the three-cluster solution. Ultimately, we decided to utilize the three-cluster solution because it was the most parsimonious and theoretically meaningful model. The internal homogeneity of each cluster was then veried using H value equations (as per Morizot and Le Blanc, 2005), where k is the number of clusters and SSD is the sum of squared distances to the mean Hk = 1 SSDwithin
k /SSDtotal sample

. The three

subgroups of participants from the three-cluster solution were internally homogeneous with excellent H values of 0.85, 0.79, and 0.81, respectively. These results ascertained that the within-cluster variances remained smaller than the variance of the total sample. In addition, a preliminary multivariate analysis of variance (MANOVA) indicated that the three-cluster solution signicantly contrasted the scores of happiness and sadness from music listening (Wilkss = 0.03, F = 716.62, P < 0.001, 2 = 0.82).1

1 In support of this three-cluster solution, its proles and their respective proportions were also reproducible using another clustering algorithm: Log-likelihood distance with Schwarzs Bayesian Criterion (for a discussion of cluster validation, see Morizot and Le Blanc, 2005).

D. Miranda, P. Gaudreau / Revue europenne de psychologie applique 61 (2011) 111

Table 3 MANOVA comparing the three music listening proles and gender on positive affects and negative affects (n = 316). Emotionally-negative listeners (n = 61) M (S.D.) Positive affect Total 30.30 (5.90) Girls 30.16 (5.93) Boys 30.50 (5.98) Negative affect Total 24.42 (5.37) Girls 23.63 (5.69) Boys 25.56 (4.75) Emotionally-limited listeners (n = 155) M (S.D.) 31.14 (6.54) 30.51 (5.96) 31.79 (7.08) 21.33 (6.49) 22.34 (6.39) 20.29 (6.47) Emotionally-positive listeners (n = 100) M (S.D.) 33.11 (7.07) 33.33 (7.20) 32.81 (6.97) 21.02 (6.26) 20.35 (5.69) 21.90 (6.92) = 1.00, F = 0.30, P > 0.05,
2

3.87*

0.02

7.10**

0.04

Main effect of music listening proles (Wilkss = 0.93, F = 5.38, P < 0.001, 2 = 0.03); main effect of gender (Wilkss music listening proles gender (Wilkss = 0.97, F = 2.05, P > 0.05, 2 = 0.01). * P < 0.05, ** P < 0.01, *** P < 0.001.

= 0.00); interaction effect of

Table 2 presents the descriptive statistics of the three-cluster solution. Standardized scores of happiness and sadness resulting from music listening (z-scores) were used in order to label the three proles more accurately. The rst cluster gathered 19.3% of adolescents (n = 61) and was named emotionally-negative listeners (ENL), because relative to the mean of the entire sample, the centroids (means) of the members of this prole were characterized by medium levels of happiness and higher levels of sadness from music listening. The second cluster grouped 49.1% of adolescents (n = 155) and was named emotionally-limited listeners (ELL), as relative to the mean of the entire sample, the centroids of the members of this prole were characterized by lower levels of both happiness and sadness from music listening. The third cluster grouped 31.6% of adolescents (n = 100) and was named emotionally-positive listeners (EPL), given that relative to the mean of the entire sample, the centroids of the members of this prole were characterized by higher levels of happiness and lower levels of sadness from music listening. Moreover, the cross-validation of the three-cluster solution was ascertained by testing it across two subsamples. As such, this threecluster solution (ENL, ELL, and EPL) was successfully replicated across the girls and boys subsamples. Table 2 presents results from 2 tests for the total sample, which indicated that the proportions of girls and boys remained similar within each of the three proles. Table 2 also presents additional 2 tests for the girls and boys subsamples, which indicated that the proportions (%) related to the three proles found in the total sample remained equivalent in the girls subsample, but not in the boys subsample. This indicated that compared to the girls, there was a smaller proportion of boys that were members of the ELL prole to the benet of a greater proportion of boys that were part of the EPL prole. Nevertheless, the three proles remained internally homogeneous with excellent H values in girls (0.82, 0.84, and 0.78) and boys (0.88, 0.78, and 0.79), respectively. Lastly, another set of preliminary MANOVA indicated that the three-cluster solution signicantly contrasted the scores of happiness and sadness from music listening in girls (Wilkss = 0.02, F = 463.45, P < 0.001, 2 = 0.85) and boys (Wilkss = 0.03, F = 313.14, P < 0.001, 2 = 0.82). In sum, these results were able to show with great consistency that the three-cluster solution (ENL, ELL, and EPL) was parsimonious, interpretable, internally homogeneous, replicable, and robust. Hence, these three clusters provided a valid and reliable basis from which to examine if three emotional proles of music listeners have different levels of emotional well-being.

in Table 3, the main effect of music listening proles (cluster) was signicant (Wilkss = 0.93, F = 5.38, P < 0.001, 2 = 0.03). However, the main effect of gender (Wilkss = 1.00, F = 0.30, P > 0.05, 2 = 0.00) and the interaction effect of music listening proles (cluster) by gender (Wilkss = 0.97, F = 2.05, P > 0.05, 2 = 0.01) were both not signicant. This indicated that gender differences were not present in our main analysis. Hence, we resumed only with the subsequent univariate analyses for music listening proles (cluster), which indicated a signicant main effect of the music listening proles for both positive affect (F = 3.87, P < 0.05) and negative affect (F = 7.10, P < 0.01). Consequently, Tukeys honestly signicant difference (Tukey HSD) post hoc tests were conducted ( = 0.05). Firstly, results indicated that EPL had higher levels of positive affect than ENL (P < 0.05) and ELL (P = 0.05). However, ENL and ELL had similar levels of positive affect (P > 0.05). Secondly, results indicated that ENL had higher levels of negative affect than ELL (P < 0.01) and EPL (P < 0.01). Nonetheless, EPL and ELL had similar levels of negative affect (P > 0.05).

3.4. Social congruence in music tastes with friends/parents and emotional well-being The secondary aim of this study focused more extensively on individual differences regarding social congruence in music tastes with friends or parents. Hence, this study also adopted a variableoriented approach which could account for associations between our socially-related variables and emotional well-being, and so, through the use of between-individuals comparisons. The variableoriented approach is the dominant paradigm in psychology, and its statistical analyses (e.g., correlations, multiple regressions) emphasize on the establishment of meaningful relationships between variables. Two moderated hierarchical regressions were performed to examine if social congruence in music tastes with friends and parents could predict either positive affect or negative affect. Moreover, because of the literature reporting gender differences in music psychology (e.g., North et al., 2000), we veried if gender was a moderator of the links between social congruence in music tastes with friends/parents and positive/negative affects. To this end, two interaction terms were entered in the equation (congruence music friends gender and congruence music parents gender), as per the approach of Cohen et al. (2003). As indicated in Table 4, in both moderated hierarchical regressions, the two interaction terms did not account for a signicant proportion of variance in positive ( R2 = 0.004, P > 0.05) and negative affect ( R2 = 0.012, P > 0.05). Gender was not a moderator of the relationships between both social congruence in music tastes with friends or with parents and either positive or negative affect. Thus, only the main effects of both social congruence in music

3.3. Emotional proles of music listeners and emotional well-being A 3 (cluster) 2 (gender) MANOVA was conducted on two dependent variables: positive affect and negative affect. As showed

D. Miranda, P. Gaudreau / Revue europenne de psychologie applique 61 (2011) 111 Table 4 Two moderated hierarchical regressions predicting either positive/negative affects levels from social congruence in musical taste with either friends/parents (n = 316). Predictors Positive affect Step 1 Gender Social congruence music friends Social congruence music parents Step 2 Congruence music friends gender Congruence music parents gender Negative affect Step 1 Gender Social congruence music friends Social congruence music parents Step 2 Congruence music friends gender Congruence music parents gender t R2

0.04 0.05 0.15 0.03 0.06

0.73 0.89 2.70** 0.54 0.97

0.026*

0.004

0.03 0.14 0.21 0.08 0.07

0.49 2.51* 3.88*** 1.44 1.23

0.063***

0.012

Variables were centered and gender was coded as girls = 1 and boys = 0. * P < 0.05, ** P < 0.01, *** P < 0.001.

tastes with friends or parents were interpreted. As can be seen in Table 4, the results indicated that higher levels of social congruence in music tastes with parents were linked to higher levels of positive affect ( = 0.15, P < 0.01), but social congruence in music tastes with friends was unrelated to positive affect ( = 0.05, P > 0.05). The results also showed that higher levels of social congruence in music tastes with friends ( = 0.14, P < 0.05) and parents ( = 0.21, P < 0.001) were both associated with lower levels of negative affect. 4. Discussion 4.1. Emotional proles of music listeners and emotional well-being The principal aim of this study was to determine if different proles (types) of emotional reactions following music listening (happiness and sadness) could characterize different levels of emotional well-being (positive and negative affects). Three proles of emotional reactions from music listening were identied in both adolescent girls and boys: ENL (medium happiness and higher sadness); ELL (lower happiness and lower sadness); EPL (higher happiness and lower sadness). In and of itself, documenting these three proles provides a pertinent contribution, as there is a limited amount of research on music listening and emotions in adolescence (Schubert and McPherson, 2006). Interestingly, we only found one prior study that has used cluster analysis on emotional reactions from music listening in adolescence, and its results were comparable to ours. Wells and Hakanen (1991) identied ve clusters from 12 emotions during music listening in adolescence, yet the robustness and homogeneity of this ve-cluster solution were not thoroughly described. Nevertheless, if we compare our clusters to those of Wells and Hakanen (1991), their angry listeners are somewhat akin to our ENL; their unemotional listeners share similitude with our ELL; and their self-assured listeners are comparable to our EPL. Their highly emotional listeners portrayed mixed feelings not found in the present study and their romantic listeners involved emotions absent from our study (e.g., love, passion). Hence, clusters of emotional reactions from music listening may vary in numbers and grow in complexity according to the number and variety of emotions studied. However, as in the present comparison, the core clusters pertaining to happiness and sadness are likely to emerge fairly consistently across studies.

The rst hypothesis was that most adolescents would be characterized as experiencing more happiness (over sadness) from music listening because such pattern could be part of their normative development. We had anticipated this outcome given that music listening generally provides positive emotional experiences and because adolescents who experience more negative emotions from music listening can have poorer mental health (Juslin et al., 2008; Martin et al., 1993; North et al., 2000; Roberts et al., 1998b; Tarrant et al., 2000). Thus, most adolescents should have been members of the EPL prole. Surprisingly, this was not conrmed by the present results, as the majority of adolescents were members of the ELL prole. This could raise an important question for developmental researchers working on music, as to whether a signicant number of adolescents are actually less emotionally reactive to (or less emotionally invested in) music listening. Accordingly, Juslin et al. (2008), who followed university students in their everyday life using palmtop electronic devices, found that music was reported as inuencing mood in 64% of everyday music listening situations not all the time. Thus, in adolescence, it could be that a frequent, intense, and greater positive (over negative) emotional reaction from music listening is not the statistical norm. That our results designated most adolescents as members of the ELL prole seems to be in accordance with the theorizing of Sloboda (2010) who has recently argued that everyday music listening may typically involve low-intensity emotional reactions. The advantage of the personoriented approach of this study emerges here, as it delineates that assuming that all (or most) adolescents are highly emotionally involved with music listening might be an overgeneralization. In brief, these results suggest that there may be different types of adolescents not necessarily a prototypical adolescent when it comes to experiencing happiness and sadness from music listening. The second hypothesis was that adolescents who experience more happiness (over sadness) from music listening would report a better emotional well-being than those experiencing more sadness (over happiness) from music listening. We had predicted this outcome because of three basic reasons: music listening usually provides positive emotions, music is very important for many adolescents, and everyday music can have moderate but cumulative emotional effects (Juslin et al., 2008; Larson, 1995; North et al., 2000; Roberts et al., 2004; Sloboda, 2010; Zillmann and Gan, 1997). Overall, the results supported this hypothesis in both girls and boys. For positive affect, EPL were higher than both ENL and ELL, whereas ENL and ELL presented similar levels. For negative affect, ENL were higher than both ELL and EPL, whereas EPL and ELL displayed similar levels. In the main, these results are compatible with the literature, which thus far suggested that adolescents who experience more negative emotions from music listening may have poorer mental health (e.g., Martin et al., 1993; Roberts et al., 1998b). From a life-span perspective, our ndings may even serve as a developmental prelude to other results linking mood regulation by music listening and psychological well-being in older adults (Laukka, 2007). Interestingly, although the majority of adolescents were members of the ELL prole, it is still the EPL prole that displayed the highest levels of emotional well-being. A frequent and intense positive (over negative) emotional reaction from music listening might not necessarily be the statistical norm. Nonetheless, for a signicant number of adolescents, an intense positive (over negative) emotional reaction from music listening may represent a valuable contributory factor to the actualization of their emotional well-being. In this sense, perhaps the EPL prole somewhat echoes the concept of strong experiences of music, which is a positive and rare peak emotional experience from music listening (Gabrielsson, 2002). Strong experiences of music can be partially interpreted

D. Miranda, P. Gaudreau / Revue europenne de psychologie applique 61 (2011) 111

within the notion of peak experiences (e.g., happiest moments, ecstatic moments, complete absorption) that can be triggered by music listening according to Abraham Maslows humanistic psychology (Gabrielsson, 2002; Maslow, 1968; North and Hargreaves, 2008). Accordingly, not only strong experiences of music were reported in adolescence, but most of them are reported by older adults as having been experienced in adolescence (Gabrielsson, 2002). Hence, adolescents in the EPL prole may be those who are more susceptible of experiencing a ow from music listening. The concept of ow refers to an optimal and intense experience involving effortless engagement, concentration, and performance in an intrinsically enjoyable activity, such as music (Bakker, 2005; Csikszentmihalyi, 1990; MacDonald et al., 2006). Thus, even if less than a third of participants were EPL, it does not signify that the role of happiness from music listening is trivial to the emotional development of adolescents. On the contrary, this provides an important nuance whereby intense happiness (over sadness) from music listening may occasionally take the form of an exceptional peak experience that can be involved in the actualization of the emotional development of many (but not most) adolescents. It is also interesting to note that although ENL have higher levels of sadness from music listening, they still experience medium levels of happiness from music listening. Indeed, happiness is a natural and particularly potent reaction from music listening (Juslin et al., 2008). Of course, it is not necessarily unusual for one to be moved to tears by the sadness (or happiness) conveyed by a musical piece (or song), especially when it is interpreted with virtuosity and/or alludes to themes deemed insightful by the listener. However, it has been documented that adolescents who expressed listening to music in order to relieve negative moods can feel worse from such behaviours in the short-term (Saarikallio and Erkkil, 2007). Thus, for some adolescents, such negative emotional reactions from music may be recurrent and less than optimal. This could be the result of the frequent venting of negative emotions, a cathartic behaviour known to be associated with lower levels of psychosocial adjustment in childhood and adolescence (Compas et al., 2001). Adolescents belonging to the ENL prole may legitimately and accurately consider that their usual reaction to music listening involves great entertainment, valuable aesthetic appreciation, and even perhaps a form of emotional catharsis. However, the resulting and recurrent negative (over positive) emotional reaction within them may unsuspectingly constitute an obstacle to the complete actualization of their emotional well-being. On the other hand, ENL represented one-fth of adolescents, numbers which incidentally parallel the 15 to 20% of adolescents that may present a given psychopathology or psychosocial problem (e.g., Shaffer et al., 1996; Roberts et al., 1998a). Thus, from the perspective of developmental psychopathology, adolescents with certain psychopathological symptoms might be the ones at greater risk of experiencing more negative (over positive) emotions from music listening. As such, their psychopathological symptoms might have generated the ENL prole. Lastly, despite that gender differences were documented for music listening in adolescence (e.g., Miranda and Claes, 2007, 2008, 2009; North et al., 2000), overall, the present person-oriented results are similarly applicable to both adolescent girls and boys. Nonetheless, if we take a step aside onto a more variable-oriented approach, the descriptive results did indicate that girls reported higher levels of happiness from music listening than boys. Indeed, this result remains compatible with variable-oriented research, which indicate that emotion regulation from music listening is greater in girls than in boys (e.g., North et al. 2000). Hence, future research could investigate if gender differences that were reported from a variable-oriented perspective to music listening in adolescence remain similar or change from a person-oriented perspective.

4.2. Social congruence in music tastes with friends/parents and emotional well-being The secondary aim of this study was to examine if there was a signicant relationship between social congruence in music tastes with friends or parents (i.e., sharing similar music tastes and having fewer conicts about music) and emotional well-being. A third hypothesis predicted that social congruence in music tastes with friends or parents would be related to more emotional well-being. Overall, the results supported our hypothesis as more social congruence in music tastes with parents was linked to more positive affect, whereas more social congruence in music tastes with friends and parents were both linked to less negative affect. Hence, this is in keeping with the mainstream literature, which indicated that good quality relationships with both peers and parents are crucial for (and part of) normative adolescent development (Hartup and Stevens, 1997; Kerr et al., 2003; Youniss and Smollar, 1985). Besides, in early childhood, Custodero et al. (2003) have reported that musical parenting (i.e., when parents played or sang music for their young children) was associated with lesser emotional distress in parents. Although this result pertained to the emotional well-being of parents, it may also signify positive musical dynamics at the family level. Such positive musical dynamic within parent-children dyads may further develop from childhood to adolescence, whereby some families keep interacting musically with each other while socially developing better emotional well-being. Hence, longitudinal studies could verify if musical parenting in childhood can predict social congruence in music between adolescents and their parents, and in turn, predict better emotional well-being in both adolescents and parents. Hence, this study goes beyond previous research about music listening and peer-relationships in adolescence (e.g., Bakagiannis and Tarrant, 2006; Miranda and Claes, 2009) by considering the role of parental relationships in the developmental importance of music listening in adolescence. As such, results suggest that not only social congruence in music tastes with friends can be associated with better emotional well-being, but also social congruence in music tastes with parents. Previous research has suggested that adolescents can experience more positive affects when listening to more unconventional and rebellious music with friends rather than with their family (Thompson and Larson, 1995). Indeed, as a resource available to adolescents, the sharing of music tastes with peers represents a powerful social tool for the development of their identity, socialization, creativity, and autonomy towards parents and other adults (e.g., Bakagiannis and Tarrant, 2006; Zillmann and Gan, 1997). In this sense, it may still be partially true that some adolescents develop autonomy from rebelling against their parents by diverging in music tastes. This could particularly be the case when adolescents consider their family relationships to be unsatisfactory, for instance, because parents are perceived as being conictual, lacking in acceptance and communication, over-controlling, conventional, boring, or simply not musical enough. However, in this study, some adolescents did not completely rebel in terms of music tastes, and perhaps they have felt better for it. Moreover, it may be that adolescents with better emotional well-being interact with parents that promote excellent relationships (e.g., communication, acceptance, reasonable control, creativity), as such that their sharing of music tastes with parents gravitates around a much broader system of positive family relationships. Thus, social congruence in music tastes with either friends or parents are probably distinct and complementary dimensions rather than merely opposing poles, an assumption that is supported by an absence of negative correlation between both variables in this study. As such, for instance, sharing music tastes may promote identity and autonomy through socialization with peers, while also

D. Miranda, P. Gaudreau / Revue europenne de psychologie applique 61 (2011) 111

promoting cultural identity and family bonding through the sharing of a cultural background transmitted by parents. In adolescence, sharing music tastes with friends still remains much more common than sharing music tastes with parents, but the latter may represent an original contributory factor to positive development, or at least, a small indicator of healthy family relationships.2 Our results encourage more studies on the inuence of parents on music listening in order to provide an original outlook to further understand the developmental role of parents in the social and emotional life of their adolescents. Besides, as media technology constantly facilitates music listening for new generations, parents (themselves former adolescents) might be more musically sophisticated (i.e., cool) in terms of emergent music than ever, and thus could increasingly enjoy sharing (and be welcomed to share) music tastes with their adolescents. Lastly, these variable-oriented results are again similarly applicable to both adolescent girls and boys. Nevertheless, there may still be some other gender differences in the various ways adolescents can use music listening behaviours to manage their social identity in the presence of other people (North et al., 2000).

Lastly, it is also important to delineate the factors that might be involved in the development of the present music listening proles (ENL, ELL, and EPL). For this matter, a useful model could be that of Hargreaves et al. (2006), which is a reciprocal feedback model of musical response that depicts interactions between the music, the listener, and the situation/context. For instance, it is possible that in the course of adolescence, the three music listening proles (ENL, ELL, and EPL) develop from complex interactions between the characteristics of songs (music and lyrics), of listeners (personality and motivation), and of situations/contexts (positive and negative social life events). 4.4. Potential practical implications: music education and prevention in adolescence Of course, much more research is needed before planning for actual interventions. Still, the results of this study suggest potential implications for music educators and mental health practitioners. Indeed, our results suggested that positive individual experiences (EPL) and positive social experiences (social congruence in music tastes with friends/parents) from music listening may potentially contribute to positive youth development (emotional well-being). Firstly, results concerning emotional reactions from music listening may provide an additional argument in favour of encouraging more music education in adolescence. This is the case given that music education can potentially promote emotional development (such as mood regulation, North and Hargreaves, 2008), and also because music education can elicit peak experiences relating to music, such as ow (e.g., Bakker, 2005; MacDonald et al., 2006). Encouraging more developmentally appropriate music education in adolescence is an important issue, as in numerous societies, music education is either limited to an elective subject or to classes primarily offered at elementary levels (North and Hargreaves, 2008). As such, emotional well-being may be an outcome targeted by music education in adolescence. Structured music education should thus be adjusted to the needs, realities, and tastes of adolescents, while taking into account that their voluntary music-related leisure activities remain developmentally signicant. As such, Larson (2000) argues that within 21th century societies that require individual autonomy, developing initiative (intrinsic motivation and deep attention) in youth is very important, and that the daily development of initiative in youth can be promoted more effectively by structured voluntary activities, such as the arts. Accordingly, the importance of informal music education (outside of schools) is nowadays catching up to that of formal music education (e.g., schools, conservatories; North and Hargreaves, 2008). Hence, it may be pertinent to encourage evidence-based, structured, and yet informal/voluntary music education programs in youth that will not only pertain to music playing, but also to music listening. Secondly, ndings about social congruence in music tastes with parents could provide new paths for evidence-based prevention programs by making them more developmentally sensitive. Within the prevention science literature, multi-components evidencedbased prevention programs in childhood and adolescence often include interventions that involve parents. For example, FASTTRACK, an effective evidence-based program for the prevention of antisocial behaviours in childhood and adolescence, considers multiple risk and protective factors, such as those relating to parenting practices (Conduct Problems Prevention Research Group, 2007). This brings us back to the recent work of Lemieux et al. (2008) who have conducted an experimental program evaluation demonstrating the effectiveness of a music-based HIV prevention-intervention for adolescents, which included evidence-based principles of the information, motivation, and behavioural skills model. Among other things, in their music-based prevention program, attention

4.3. Research limitations: bi-directional interpretations for correlations The self-reported measures pertaining to music listening were preliminary in nature. For instance, the minimal number of items (two-item scales) for social congruence in music tastes with friends/parents has probably limited reliability. Our very short measure of emotional states following music listening remained exploratory and it could not take into account the social context (e.g., alone, with friends, or with family) nor the motivation for listening to music (e.g., for ambiance, party, identity). As McFerran (2010) pointed out, adolescents can give importance to whether they are getting expected social reactions through music. The extent to which adolescents are fullling different social expectations through different music listening behaviours may impact their emotions in various ways that are not specically captured by our short measures. Moreover, it should be mentioned again that our single-item measure of sadness resulting from music listening may have lacked specicity: I feel unhappy after listening to my favourite music. . .. This label should have been interpreted as sadness, but may have overlapped with other negative emotions (e.g., frustration). Also, the theoretical stance of this study remained speculative in suggesting that experiencing more happiness (as opposed to sadness) from music listening may be a factor, among others, that could further develop the emotional wellbeing of adolescents (music emotional well-being). Indeed, the present cross-sectional and correlational results do not allow examining direction, prediction, or causality between variables. As such, it may also be that adolescents who present better emotional well-being are more prone to experiencing and beneting from positive emotions as a result of music listening (emotional well-being music). Moreover, the relation may be reciprocal (bi-directional) with retroactive loops that can only be identied within a longitudinal design (e.g., music emotional well-being music emotional well-being music emotional well-being). In all cases, researchers and clinicians should consider that resilient and non-resilient adolescents may use music listening in different ways (McFerran, 2010).

2 Indeed, the mean level of social congruence in music tastes with friends was higher than that of social congruence in music tastes with parents (t = 24.96, P < 0.001).

10

D. Miranda, P. Gaudreau / Revue europenne de psychologie applique 61 (2011) 111 Gaudreau, P., Blondin, J.P., 2004. Different athletes cope differently during a sport competition: a cluster analysis of coping. Personality and Individual Differences 36, 18651877. Gaudreau, P., Sanchez, X., Blondin, J.P., 2006. Positive and negative affective states in a performance-related setting: testing the factorial validity of the PANAS across two samples of French-Canadian participants. European Journal of Psychological Assessment 22, 240249. Hargreaves, D.J., 1986. The Developmental Psychology of Music. Cambridge University Press, Cambridge. Hargreaves, D.J., North, A.C., Tarrant, M., 2006. Musical preference and taste in childhood and adolescence. In: McPherson, G.E. (Ed.), The Child as Musician. Oxford University Press, Oxford, pp. 135154. Harmon-Jones, E., Harmon-Jones, C., Abramson, L., Peterson, C.K., 2009. PANAS positive activation is associated with anger. Emotion 9, 183196. Hartup, W.W., Stevens, N., 1997. Friendships and adaptation in the life course. Psychological Bulletin 121, 355370. Hunter, P.G., Schellenberg, E.G., Schimmack, U., 2008. Mixed affective responses to music with conicting cues. Cognition and Emotion 22, 327352. Juslin, P.N., Liljestrm, S., Vstfjll, D., Barradas, G., Silva, A., 2008. An experience sampling study of emotional reactions to music: listener, music, and situation. Emotion 8, 668683. Juslin, P.N., Sloboda, J.A., 2010. Handbook of Music and Emotion: Theory, Research, Applications. Oxford University Press, New York. Kallinen, K., 2005. Emotional ratings of music excerpts in the western art music repertoire and their self-organization in the Kohonen neural network. Psychology of Music 33, 373393. Kerr, M., Stattin, H., Biesecker, G., Ferrer-Wreder, L., 2003. Relationships with parents and peers in adolescence. In: Lerner, R.M., Easterbrooks, M.A., Mistry, J., Weiner, I.B. (Eds.), Handbook of Psychology (vol. 6): Developmental Psychology. Wiley, Hoboken, pp. 395419. Koestner, R., Lekes, N., Powers, T.A., Chicoine, E., 2002. Attaining personal goals: self-concordance plus implementation intentions equals success. Journal of Personality and Social Psychology 83, 231244. Lacourse, E., 1999. Alination, musique heavy metal et risque suicidaire chez les adolescents [Alienation, heavy metal music and suicidal risks in adolescents]. Universit de Montral, Montreal, Canada (unpublished doctoral dissertation). Larson, R., 1995. Secrets in the bedroom: adolescents private use of media. Journal of Youth and Adolescence 24, 535550. Larson, R.W., 2000. Toward a psychology of positive youth development. American Psychologist 55, 170183. Laukka, P., 2007. Uses of music and psychological well-being among elderly. Journal of Happiness Studies 8, 215241. Lemieux, A.F., Fisher, J.D., Pratto, F., 2008. A music-based HIV prevention intervention for urban adolescents. Health Psychology 27, 349357. MacDonald, R., Byrne, C., Carlton, L., 2006. Creativity and ow in musical composition: an empirical investigation. Psychology of Music 34, 292306. Martin, G., Clarke, M., Pearce, C., 1993. Adolescent suicide: music preference as an indicator of vulnerability. Journal of the American Child & Adolescent Psychiatry 32, 530535. Maslow, A.H., 1968. Toward a Psychology of Being, second ed. Van Nostrand Reinhold, New York. McFerran, K., 2010. Adolescents, Music and Music Therapy: Methods and Techniques for Clinicians, Educators, and Students. Jessica Kingsley Publishers, London. McPherson, G.E., 2006. The Child as Musician: A Handbook of Musical Development. Oxford University Press, Oxford. McPherson, G.E., 2009. The role of parents in childrens musical development. Psychology of Music 37, 91110. Miranda, D., Claes, M., 2007. Musical preferences and depression in adolescence. International Journal of Adolescence and Youth 13, 285309. Miranda, D., Claes, M., 2008. Personality traits, music preferences and depression in adolescence. International Journal of Adolescence and Youth 14, 277 298. Miranda, D., Claes, M., 2009. Music listening, coping, peer afliation and depression in adolescence. Psychology of Music 37, 215233. Morizot, J., Le Blanc, M., 2005. Searching for a developmental typology of personality and its relations to antisocial behavior: a longitudinal study of a representative sample of men. Journal of Personality 73, 139182. Mulder, J., ter Bogt, T., Raaijmakers, Q., Vollebergh, W., 2007. Music taste groups and problem behaviour. Journal of Youth and Adolescence 36, 313324. North, A.C., Hargreaves, D.J., 2008. The Social and Applied Psychology of Music. Oxford University Press, Oxford. North, A.C., Hargreaves, D.J., ONeill, S.A., 2000. The importance of music to adolescents. British Journal of Educational Psychology 70, 255272. Peretz, I., Zatorre, R., 2003. The Cognitive Neuroscience of Music. Oxford University Press, New York. Rentfrow, P.J., Gosling, S.D., 2003. The do re mis of everyday life: the structure and personality correlates of music preferences. Journal of Personality and Social Psychology 84, 12361256. Roberts, R.E., Attkisson, C.C., Rosenblatt, A., 1998a. Prevalence of psychopathology among children and adolescents. American Journal of Psychiatry 155, 715725. Roberts, K.R., Dimsdale, J., East, P., Friedman, L., 1998b. Adolescent emotional response to music and its relationship to risk-taking behaviors. Journal of Adolescent Health 23, 4954. Roberts, D.F., Henriksen, L., Foehr, U.G., 2004. Adolescents and media. In: Lerner, R.M., Steinberg, L. (Eds.), Handbook of Adolescent Psychology, second ed. Wiley, Hoboken, pp. 487521.

was given to the importance of peers in music-related activities in adolescence, which was in line with the current literature on music in adolescence. However, the present study can suggest that exploring the importance of parental relationships in the developmental signicance of music listening in adolescence could also provide a potential adjunct to such developmentally-relevant and effective prevention programs. For instance, music could be used to promote family bonding and communication between parents and adolescents. Lastly, our exploratory ndings may eventually have some implications for music therapy in youth. Hitherto, music therapy for adolescents has usually focused on individual rather than group work despite the social importance of music (McFerran, 2010). Accordingly, the present ndings may suggest that both peers and parents should be considered in future adolescent research on group music therapy within educational and community settings. Conict of interest statement None. Acknowledgments This study was supported in part by a scholarship for postdoctoral research from the Fonds qubcois de la recherche sur la socit et la culture (FQRSC) awarded to the rst author. This study was also supported in part by a research grant from the Social Sciences and Humanities Research Council of Canada (SSHRC) awarded to the second author. The authors also thank Marie-Claude Fecteau and Amanda Thompson for their comments on an earlier version of this article. References
Aldenderfer, M.S., Blasheld, R.K., 1984. Cluster Analysis. Sage, Beverly Hills, CA. Bakagiannis, S., Tarrant, M., 2006. Can music bring people together? Effects of shared musical preference on intergroup bias in adolescence. Scandinavian Journal of Psychology 47, 129136. Bakker, A.B., 2005. Flow among music teachers and their students: the crossover of peak experiences. Journal of Vocational Behavior 66, 2644. Brown, B.B., Klute, C., 2003. Friendships, cliques, and crowds. In: Adams, G.R., Berzonsky, M.D. (Eds.), Blackwell Handbook of Adolescence. Blackwell Publishing, Malden, pp. 330348. Brown, B.B., Larson, J., 2009. Peer relationships in adolescence. In: Lerner, R.M., Steinberg, L. (Eds.), Handbook of Adolescent Psychology, third ed. (volume 2): Contextual Inuences on Adolescent Development. Wiley, Hoboken, NJ, pp. 74103. Busseri, M.A., Sadava, S., Molnar, D., DeCourville, N., 2009. A person-centered approach to subjective well-being. Journal of Happiness Studies 10, 161181. Cohen, J., Cohen, P., West, S.G., Aiken, L.S., 2003. Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences, third ed. Lawrence Erlbaum, Mahwah, NJ. Compas, B.E., Connor-Smith, J.K., Saltzman, H., Harding-Thomsen, A., Wadsworth, M.E., 2001. Coping with stress during childhood and adolescence: problems, progress, and potential in theory and research. Psychological Bulletin 127, 87127. Conduct Problems Prevention Research Group, 2007. Fast track randomized controlled trial to prevent externalizing psychiatric disorders: ndings from grades 3 to 9. Journal of the American Academy of Child and Adolescent Psychiatry 46, 12501262. Csikszentmihalyi, M., 1990. Flow: The Psychology of Optimal Experience. Harper Collins, New York. Custodero, L.A., Britto, P.R., Brooks-Gunn, J., 2003. Musical lives: a collective portrait of American parents and their young children. Applied Developmental Psychology 24, 553572. Delsing, M.J.M.H., ter Bogt, T.F.M., Engels, R.C.M.E., Meeus, W.H.J., 2008. Adolescents music preferences and personality characteristics. European Journal of Personality 22, 109130. Diener, E., 2000. Subjective well-being: the science of happiness and a proposal for a national index. American Psychologist 55, 3443. Gabrielsson, A., 2002. Old peoples remembrance of strong experiences related to music. Psychomusicology 18, 103122. Gaudreau, P., Amiot, C.E., Vallerand, R.J., 2009. Trajectories of affective states in adolescent hockey players: turning point and motivational antecedents. Developmental Psychology 45, 307319.

D. Miranda, P. Gaudreau / Revue europenne de psychologie applique 61 (2011) 111 Rogers, C.R., 1961. On Becoming a Person: A Therapists View of Psychotherapy. Mariner Books/Houghton Mifin Company, New York. Saarikallio, A., Erkkil, J., 2007. The role of music in adolescents mood regulation. Psychology of Music 35, 88109. Schubert, E., McPherson, G.E., 2006. The perception of emotion in music. In: McPherson, G.E. (Ed.), The Child as Musician. Oxford University Press, Oxford, pp. 193212. Selfhout, M.H.W., Delsing, M.J.M.H., ter Bogt, T.F.M., Meeus, W.H.J., 2008. Heavy metal and hip-hop style preferences and externalizing problem behavior: a two-wave longitudinal study. Youth & Society 39, 435452. Selfhout, M.H.W., Branje, S.J.T., ter Bogt, T.F.M., Meeus, W.H.J., 2009. The role of music preferences in early adolescents friendship formation and stability. Journal of Adolescence 32, 95107. Seligman, E.P.M., Csikszentmihalyi, M., 2000. Positive psychology: an introduction. American Psychologist 55, 514. Shaffer, D., Fisher, P., Dulcan, M.K., Davies, M., Piacentini, J., Schwab-Stone, M.E., et al., 1996. The NIMH diagnostic interview schedule for children version 2.3 (DISC2.3): description, acceptability, prevalence rates, and performance in the MECA study. Journal of the American Academy of Child and Adolescent Psychiatry 35, 865877. Sloboda, J.A., 2010. Music in everyday life: the role of emotions. In: Juslin, P.N., Sloboda, J.A. (Eds.), Handbook of Music and Emotion: Theory, Research, Applications. Oxford University Press, New York, pp. 493514.

11

Tarrant, M., North, A.C., Hargreaves, D.J., 2000. English and American adolescents reasons for listening to music. Psychology of Music 28, 166173. Thompson, R.L., Larson, R., 1995. Social context and the subjective experience of different types of rock music. Journal of Youth and Adolescence 24, 731744. Watson, D., Clark, L.A., 1997. Measurement and mismeasurement of mood: recurrent and emergent issues. Journal of Personality Assessment 68, 267296. Watson, D., Clark, L.A., Tellegen, A., 1988. Development and validation of brief measures of positive and negative affect: the PANAS scales. Journal of Personality and Social Psychology 54, 10631070. Watson, D., Tellegen, A., 1985. Toward a consensual structure of mood. Psychological Bulletin 98, 219235. Wells, A., Hakanen, E.A., 1991. The emotional use of popular music by adolescents. Journalism Quarterly 68, 445454. Wilson, K., Gullone, E., 1999. The relationship between personality and affect over the lifespan. Personality and Individual Differences 27, 11411156. Youniss, J., Smollar, J., 1985. Adolescent Relations with Mothers, Fathers, and Friends. University of Chicago Press, Chicago. Zillmann, D., Gan, S., 1997. Musical taste in adolescence. In: Hargreaves, J., North, A.C. (Eds.), The Social Psychology of Music. Oxford University Press, New York, pp. 161187.